随着智能变电站的建设,继电保护原理的实现技术发生了巨大的变化.传统验证性、原理性的实验平台难以适应工程现场继电保护专业技术的应用.因此,迫切需要对现有的教学实验基础条件进行改造和建设.因此,本文阐述四川大学新建继电保护综合实验室的建设思路和方案.实验室以现场电力系统智能变电站建设标准为参考,引进工业化的微机继电保护装置及监控系统,最大限度地模拟实际工程应用,以期达到学生在校实验与工程操作无缝对接的教学目标.

《电力系统不确定性理论与测度》是电力专业里一门交叉型课程,根据电力系统课程的独特性,结合现有教学环境,按照课程教学目的,文章根据课程教学中出现的缺点与不足,从教学内容、教学理念和教学方式等方面进行了改革.通过引入案例教学,激发学生对电力系统不确定性理论深入了解的热情,使学生更加充分理解课程内容,同时培养学生的分析解决问题能力、工程实践能力,整体上提高学生的综合素质.

为准备识别复杂电能质量扰动类型,提出一种基于条件互信息平均最优化(avg-CMIM)特征选择法与Adaboost动态集成分类器的电能质量复合扰动分类策略.首先基于条件互信息,提出准确衡量特征与扰动类别相关性、特征集内部目标导向冗余性的评价准则,得到不同扰动标签相匹配的最优分类特征集.再利用Adaboost分类器进行动态增强学习,对未知样本进行标签识别,通过组合标签结果确定复合扰动的组成成分,实现电能质量复合扰动的识别.仿真结果表明,在不同程度噪音下,该方法能够高效准确地识别电压暂升、电压暂降、电压短时中断、谐波、脉冲暂态和振荡暂态等单一扰动和其组合成的复合扰动,并通过实测数据验证了方法的正确性和可行性.

为了适应新的严峻国际竞争和新经济发展,着力推进新工科建设,电气工程专业从电能的生产与传输出发,围绕电力装备的研发、跨学科的综合、新能源发电、智能电网和能源互联网展开工程实习实训教学内容探索.实习实训从设备的认知到操作、检修与巡检,引导研发新产品.通过与客户和能源主题之间的互联协调,锻炼学生跨学科和优质服务的实训,培养综合高素质的人才.经过实践教学,取得良好的效果.

基于攻击者视角分析电力信息物理融合系统的脆弱性有助于决策者辨识系统脆弱环节和潜在威胁,为开展针对性防御奠定基础。结合复杂网络理论,建立了考虑信息物理交互的电力信息物理融合系统一体化模型;在直流潮流模型下,研究了信息层元件失效对物理层故障传播的影响;从攻击者角度,建立了4种攻击模式,从物理层和信息层结构与功能属性出发,建立了更为全面的融合系统脆弱性指标,并对2种信息网拓扑结构下融合系统的脆弱性进行了分析。仿真结果表明,物理层负载水平和信息层拓扑结构是影响融合系统脆弱性的重要因素;相比随机攻击,蓄意攻击下融合系统更脆弱,协同攻击下尤为脆弱;高加权介数和高加权度数信息节点对维持信息层功能起关键作用;信息物理协同破坏效应严重恶化了系统的性能。

考虑信息物理交互研究电力–信息耦合网络脆弱性及其改善策略具有重要意义。在直流潮流模型下,建立考虑信息物理交互的电力–信息耦合网络故障模型,研究信息网元件失效对电力网连锁故障的影响,分析由此引发的电力–信息耦合网络脆弱性;为深入揭示耦合网络脆弱性,在度数和介数指标基础上,采用映射加权方式计及监控功能属性和信息物理耦合的影响,提出了关键度指标辨识对耦合网络脆弱性具有重要影响的信息网关键环节;为改善耦合网络脆弱性,从强化关键环节防护和降低网络对关键环节依赖性角度,在元件和网络层面分别提出信息网关键环节保护和高权重电力节点网间加边策略。节点和边攻击的仿真结果表明,耦合网络面对随机攻击呈现较高鲁棒性,面对蓄意攻击呈现较高脆弱性。高关键度信息元件是影响耦合网络脆弱性的信息网关键环节,不同场景中耦合网络在针对这些元件的攻击下均呈现高脆弱性。保护高关键度信息元件以及对高权重电力节点增加耦合边可有效改善耦合网络脆弱性。

The growth of electricity demand is closely related to economic structure.This paper presents an empirical analysis model of electricity demand in China by considering the existence of spatial effects from regional development point of view.According to spatial panel data of electricity consumption in 30 provincial areas from 1999 to 2014,the influence of such economic structure factors as industrial structure,regional structure,distribution structure and energy structure on electricity demand and its spatial effect are analyzed.The spatial effects are set,tested and optimally selected by combing "from specific to general" method with "from general to specific" method,and a bias correction is carried out for fixed effect model.On the basis of overall analysis,a comparison analysis on the direct effects and the indirect effects of explained variables in different regions is made.The empirical analysis results show that the economic structure has an important influence on electricity demand,provincial electricity demand has obvious spatial interaction effects during the sample period,selected explained variables will change the electricity demand of sample provinces and its neighboring provinces,and the direct effect and indirect effect of explained variables in different regions are different obviously.In addition,the trends of electricity demand during the 13th Five-Year Plan period of China are analyzed and the proposal for the regional planning is put forward.

Online assessment of power grid cascading failure lies in acquiring physical models and real-time data.A method based on CIM/E for power grid cascading failure online assessment is proposed in this paper.In view of the characteristics of power system physical models and data,on the basis of applying CIM/E to power system cascading failure online assessment,document check,data extraction and data processing of CIM/E and island recognition are researched.Then a complete cascading failure online assessment scheme of regional system is presented combining the above method and credibility measurement.Through system development and application in southwest of China practical regional system,and both the correctness and the effectiveness of the proposed application were verified.The method is able to timely and accurately provide decision-making basis for the dispatching personnel.

The functions of information system should be realized through the information network.Different damaged degrees of information network correspond to different functional states of the information system, which affects the emergency response capacity of the power system.Research on the impact of information network damage on cascading failures of power grid is an important topic for ensuring safe operation of the modern power network.A model of cascading failures in a power grid considering the impact of information network damage is built.Damage indices of the structure of information network and functions of information system are proposed to analyze the impact of different damaged degrees of information network on cascading failures.From the perspective of regional distribution and individual importance of failure edges, the impact of information network damage on cascading failures is investigated with several failure modes.Results of an IEEE 30-bus system show that the ability to restrain the propagation of cascading failures is weakened by the damaged information network, which leads to the spread of cascading failures.When the power grid is under a heavy load level, the impact of information network damage on cascading failures is more apparent.Differences in the scale of cascading failures are obvious among different failure modes, and structural vulnerability of information network is an important factor affecting the spread of cascading failures of power grids.

A method based on the interaction between the primary and secondary systems of power grid is introduced for analyzing the mechanism of cascading failure evolution.The concept of interaction is based on the physical mutual impact between the primary and secondary systems and its function during the evolution of cascading failure is analyzed.A concept of protection trip severity is applied to characterize this interac tion and a variable weight model of primary system is established.A weighted topology model based on the interaction is built and the network efficiency index is adopted to analyze the mechanism of cascading failure evolution.Simulative results show that,power system becomes more vulnerable under this interaction;the bigger the operating capacity c(o)efficient is,the greater the cascading failure impacts on network efficiency;the bigger the operating limit coefficient is,the stronger the robustness of power grid against cascading failure becomes.

With steady liberalization of the electricity market,competition has been introduced into the sale of electricity.Providing quality customers with quality power supply service appears to be particularly important.The voltage sag insurance mechanism is exactly an effective way to improve users' satisfaction.Based on an analysis of the loss risk characteristics of voltage sag,the control type and financial type loss risk management modes are compared and discussed.Three voltage sag insurance mechanisms of financial risk management mode are set.By introducing the grey system theory and making comprehensive quantitative assessment of different insurance mechanisms through synthesizing the views of different entities,the most suitable and feasible insurance mechanism for the loss risk of voltage sag has been obtained.Finally,according to an empirical analysis of three high-tech customers in a high-tech park in China,the applicability and rationality of the method is proved.

Abstract:In order to improve operators’ situation awareness of dealing with loop closing scenarios, a fast simulation method of distribution network loop closing based on compensation method and nodal equation was proposed to simulate the influence of loop closing on distribution network rapidly. The change of loop closing electrical quantities was depicted by compensation method, and loop closing simulation formulas based on incidence matrix and admittance matrix were derived with nodal equation to quantitatively describe the influence of loop closing on distribution network. Only running state and network parameters before loop closing were needed to calculate the state after loop closing, and there was no need to recalculate power flow. The PG&E 69-node system and a practical distribution network were simulated, and the results proved the validity and practicality of the proposed method.

本文设计实现了全开放式新能源实验平台,综合了设计、测试、实验与组网功能,可实现控制算法级至系统拓扑级的自主实验与开发.该平台上可发展验证型、设计型、综合型等与新能源相关的四大类、二十余项实验项目,满足本科及研究生课堂教学、课程设计、毕业设计及大学生创新实践训练等多层次教学需要.实践表明,该实验平台在引领学生在电气及新能源领域的创新和实践过程中发挥了重要作用.

Protection failures have serious threat to power grid security. It is a typical dynamic equilibrium characteristic when line parameters are revised to mitigate cascading failures after the fault occurs. Based on complex network theory a method of risk assessment, considering protection failures and power grid dynamic equilibrium characteristics, is proposed. Firstly, this paper defines a concept of protection failure severity, measuring the impact of protection failure on power grid security, based on hidden failures. Secondly, a variable weighted model, combined with protection failure severity, is established to characterize the dynamic of power grid. Then, risk theory is introduced to assess power grid security from the point of complex network theory. Finally, the New England 39-bus system is taken as an example and the validity of proposed method is verified.

针对电力系统暂态稳定评估中稳定样本与不稳定样本误分类代价不同的特点,提出一种基于代价敏感极端学习机的电力系统暂态稳定评估方法.该方法在现有极端学习机的基础上,引入误分类代价的概念,以误分类代价最小为目标,构造代价敏感极端学习机,克服了现有极端学习机应用于暂态稳定评估时只追求高的分类准确率而忽略不稳定样本漏报率的缺点.新英格兰39节点系统和IEEE 145节点系统的仿真结果表明,所提方法的评估结果更倾向于将样本划分为误分类代价大的不稳定样本,以减小总的误分类代价.通过调整误分类代价矩阵,不仅可以使漏报率降为0,还能使稳定样本的误报率维持在较低的水平,保证了评估结果的可靠性.

电网关键线路对连锁故障的发展起到推波助澜的作用.本文提出了一种综合计及电网结构和运行状态的识别方法.利用线路介数量化线路在电网结构中的关键程度;基于风险理论和可信性理论,求取线路断开后给系统带来的风险,得到线路在运行状态上的关键程度.最后提出了考虑结构和运行状态的综合关键指标,用以识别关键线路.对IEEE-RTS 24节点系统仿真,证明了本文方法能够有效识别出对大停电产生重要影响的一系列关键线路.

Loop closing scenarios change with distributed generations connected to distribution network, which puts forward new requirement for the calculation of loop closing power flow. A loop closing power flow calculation algorithm of distribution network with PV nodes is proposed. For the weak ability of back-forward sweep method to deal with PV nodes and looped network, the improved sensitivity matrix is used to realize the processing of PV nodes with power correction, and based on the superposition principle, the calculation of loop closing power flow is realized with loop closing power flow compensation method. The algorithm performance is verified by IEEE 33-node test system, results show that the proposed algorithm can effectively solve the calculation problem of loop closing power flow in distribution network with PV nodes.

With the development of power grid , the inter-area low-frequency oscillations become one of the increasing con-cerns in the stable operation of power system .A method using wavelet soft threshold denoising technology is proposed .Firstly, the low-frequency oscillation data of power system is preprocessed , and then the characteristics of low -frequency oscillation signal are extracted by stochastic subspace algorithm .This method identifies the low-frequency oscillation and its characteris-tic parameters based on on -line measurement data directly , and effectively overcomes the defects that Prony algorithm , auto-regressive and moving average ( ARMA) algorithm and Hilb-Huang transform algorithm are influenced by noise and actual order number of the system as well as the shortcomings that it is difficult for single stochastic subspace to deal with nonlinear and non-stationary oscillation signals .The feasibility of applying the proposed method to the analysis on low -frequency os-cillation of power system is verified by numerical simulation and instance analysis .

In large-scale inter-connected power grids,low frequency oscillation has become a serious threat to the wide-area safe and stable operation of power systems.The application of WAMS(Wide Area Measurement System) opens the way to the online analysis of low frequency oscillation,which can be achieved through the online identification and monitoring,and applying the appropriate damping Control.This paper reviews the online analysis methods based on ambient excitation,emphatically,their fundamental principle,characteristics and application.Finally,some existing problems and further research direction are discussed.

The current stochastic assessment method usually adopts the probability measure to quantify the uncertainty of cascading failures. There is a contradiction between simulation precise and samplings because of the rarity of cascading failures. The cascading failure event was treated as a fuzzy one based on the united principle of fuzziness and randomness in this paper. The risk evaluation measure system, including the credibility measure, total load loss and risk measure, was introduced based on the credibility theory. The hidden failure model of protection systems was taken as the propagation mechanism of cascading failures, and then the basic framework and mathematical model of identification and risk assessment of cascading failures were proposed through historical statistical information or artificial contingency sets. On the simulation of the WSCC nine-bus and IEEE thirty-bus test system, compared with the stochastic assessment method, the results have shown that the proposed method is accurate, valid, little dependent on samplings and robust, and it is suitable for practical engineering applications.